@@ -19,9 +19,7 @@ This software is made available under the MIT License, see LICENSE
== SYNOPSIS:
- tagger = Brill::Tagger.new( File.join(File.dirname(__FILE__),"LEXICON"),
- File.join(File.dirname(__FILE__),"LEXICALRULEFILE"),
- File.join(File.dirname(__FILE__),"CONTEXTUALRULEFILE") )
+ tagger = Brill::Tagger.new
== INSTALL:
README.txt
@@ -18,10 +18,12 @@ module Brill
# given a body of text return a list of adjectives
def adjectives( text )
+ tag(text).select{|t| t.last == 'JJ' }
end
# given a body of text return a list of nouns
def nouns( text )
+ tag(text).select{|t| t.last.match(/NN/) }
end
# returns similar results as tag, but further reduced by only selecting nouns
@@ -92,6 +94,7 @@ module Brill
# returns an array like [[token,tag],[token,tag]...[token,tag]]
#
def tag( text )
+ # XXX: the list of contractions is much larger then this... find'em
text = text.gsub(/dont/,"don't").gsub(/Dont/,"Don't")
text = text.gsub(/youre/,"you're")
tokens = Brill::Tagger.tokenize( text )
lib/brill/tagger.rb
@@ -139,6 +139,11 @@ TMZ.com: Britney celebrated getting overnights with her kids by going on a wild
puts results.inspect
end
+ def test_adjectives
+ results = tagger.adjectives("So happy i get to bring my baby boy home tomorrow. Hospital tv is horrible, ten channels no one watches")
+ assert_equal [["happy", "JJ"], ["horrible", "JJ"]], results
+ end
+
private
def tagger
$rtagger
test/test_rule_tagger.rb
@@ -175,9 +175,7 @@ gem install rbtagger
<p><pre class='syntax'>
<span class="ident">require</span> <span class="punct">'</span><span class="string">rbtagger</span><span class="punct">'</span>
-<span class="ident">tagger</span> <span class="punct">=</span> <span class="constant">Brill</span><span class="punct">::</span><span class="constant">Tagger</span><span class="punct">.</span><span class="ident">new</span><span class="punct">(</span> <span class="constant">File</span><span class="punct">.</span><span class="ident">join</span><span class="punct">(</span><span class="constant">File</span><span class="punct">.</span><span class="ident">dirname</span><span class="punct">(</span><span class="constant">__FILE__</span><span class="punct">),"</span><span class="string">LEXICON</span><span class="punct">"),</span>
- <span class="constant">File</span><span class="punct">.</span><span class="ident">join</span><span class="punct">(</span><span class="constant">File</span><span class="punct">.</span><span class="ident">dirname</span><span class="punct">(</span><span class="constant">__FILE__</span><span class="punct">),"</span><span class="string">LEXICALRULEFILE</span><span class="punct">"),</span>
- <span class="constant">File</span><span class="punct">.</span><span class="ident">join</span><span class="punct">(</span><span class="constant">File</span><span class="punct">.</span><span class="ident">dirname</span><span class="punct">(</span><span class="constant">__FILE__</span><span class="punct">),"</span><span class="string">CONTEXTUALRULEFILE</span><span class="punct">")</span> <span class="punct">)</span>
+<span class="ident">tagger</span> <span class="punct">=</span> <span class="constant">Brill</span><span class="punct">::</span><span class="constant">Tagger</span><span class="punct">.</span><span class="ident">new</span>
<span class="ident">docs</span><span class="punct">.</span><span class="ident">each</span> <span class="keyword">do</span><span class="punct">|</span><span class="ident">doc</span><span class="punct">|</span>
<span class="ident">tagger</span><span class="punct">.</span><span class="ident">tag</span><span class="punct">(</span> <span class="constant">File</span><span class="punct">.</span><span class="ident">read</span><span class="punct">(</span> <span class="ident">doc</span> <span class="punct">)</span> <span class="punct">)</span>
<span class="keyword">end</span>
@@ -210,7 +208,7 @@ rake install_gem</pre>
<h2>Contact</h2>
<p>Comments are welcome. Send an email to <a href="mailto:rb-brill-tagger@googlegroups.com">Todd A. Fisher</a> email via the <a href="http://groups.google.com/group/rb-brill-tagger">forum</a></p>
<p class="coda">
- <a href="http://xullicious.blogspot.com/">Todd A. Fisher</a>, 23rd June 2008<br>
+ <a href="http://xullicious.blogspot.com/">Todd A. Fisher</a>, 21st May 2009<br>
Theme extended from <a href="http://rb2js.rubyforge.org/">Paul Battley</a>
</p>
</div>
website/index.html
7968881972ddbc5d285306cd1e1a88a5c2e50fb6
Todd Fisher
todd.fisher@gmail.com
http://github.com/taf2/rb-brill-tagger/commit/7194f3f5736832feca66432d799ec830a8d52dc1
7194f3f5736832feca66432d799ec830a8d52dc1
2009-05-21T19:41:11-07:00
2009-05-21T19:41:11-07:00
easy extract nouns and adjectives
2cefc777f95a1ece6820d1c59491123cc01ffbf1
Todd Fisher
todd.fisher@gmail.com